Origin of the gamma rays from the Galactic Center

The region surrounding the center of the Milky Way is both astrophysically rich and complex, and is predicted to contain very high densities of dark matter. Utilizing three years of data from the Fermi Gamma Ray Space Telescope (and the recently available Pass 7 ultraclean event class), we study the morphology and spectrum of the gamma ray emission from this region and find evidence of a spatially extended component which peaks at energies between 300 MeV and 10 GeV. We compare our results to those reported by other groups and find good agreement. The extended emission could potentially originate from either the annihilations of dark matter particles in the inner galaxy, or from the collisions of high-energy protons that are accelerated by the Milky Way’s supermassive black hole with gas. If interpreted as dark matter annihilation products, the emission spectrum favors dark matter particles with a mass in the range of 7–12 GeV (if annihilating dominantly to leptons) or 25–45 GeV (if annihilating dominantly to hadronic final states). The intensity of the emission corresponds to a dark matter annihilation cross section consistent with that required to generate the observed cosmological abundance in the early Universe ($\sigma v\sim 3\times {10}^{-26}{\mathrm{cm}}^3/\mathrm{s}$). We also present conservative limits on the dark matter annihilation cross section which are at least as stringent as those derived from other observations.

Figure 1

Contour maps of the gamma ray flux from the region surrounding the Galactic Center, as observed by the Fermi Gamma Ray Space Telescope. The left frames show the raw maps, while the center and right frames show the maps after subtracting known sources (not including the central source), and known sources plus emission from cosmic ray interactions with gas in the Galactic Disk, respectively. All maps have been smoothed over a scale of 0.5 degrees. See text for more details.

Figure 2

The observed gamma ray flux (after subtracting known sources) averaged over $7°<|l|<10°$ as a function of galactic latitude (solid), compared to that predicted from the line-of-sight gas density (dashes). See text for details.

Figure 3

The spectrum of the residual emission from the inner 5 degrees surrounding the Galactic Center, after subtracting the known sources and line-of-sight gas templates.

Figure 4

Fits for the spectrum of the central emission, assuming a pointlike source morphology, from the previous work of three different groups (see the left frame of Fig. 14 in Ref. 5, Fig. 2 in Ref. 7, and Fig. 3 in Ref. 6). Despite the different analysis approaches taken, these fits are all in reasonable agreement. The dashed line is the broken power-law fit to this spectrum as presented in Ref. 6.

Figure 5

A comparison of the total residual emission found in this study (black) with the spectra of pointlike emission (red) and extended emission (blue) (as in the case of annihilating dark matter with ${\rho }_{\mathrm{DM}}\propto r^{-1.34}$) as presented in Ref. 7 (Figs. 2, 5). This comparison supports our finding that this residual emission below $\sim 300\mathrm{MeV}$ is consistent with a pointlike source origin, while much of the emission at higher energies is indeed spatially extended.

Figure 6

The range of dark matter masses and annihilation cross sections for which dark matter annihilations can account for the majority of the observed residual emission between 300 MeV and 10 GeV, for three choices of annihilation channels (“leptons” denotes equal fractions to $e^{+}{e}^{-}$, ${\mu }^{+}{\mu }^{-}$ and ${\tau }^{+}{\tau }^{-}$). Also shown for comparison is the annihilation cross section predicted for a simple thermal relic ($\sigma v=3\times {10}^{-26}{\mathrm{cm}}^3/\mathrm{s}$). Note that there is a factor of a few uncertainty in the annihilation cross section, corresponding to the overall dark matter density and distribution. See text for details.

Figure 7

Examples illustrating how dark matter annihilations and astrophysical sources could combine to make up the observed residual emission surrounding the Galactic Center. In the upper left frame, we show results for a 10 GeV dark matter particle with an annihilation cross section of $\sigma v=7\times {10}^{-27}{\mathrm{cm}}^3/\mathrm{s}$ and which annihilates only to leptons ($e^{+}{e}^{-}$, ${\mu }^{+}{\mu }^{-}$ and ${\tau }^{+}{\tau }^{-}$, $1/3$ of the time to each). In the upper right frame, we show the same case, but with 10% of the annihilations proceeding to $b\overline{b}$. In the lower frame, we show results for a 30 GeV dark matter particle annihilating to $b\overline{b}$ with an annihilation cross section of $\sigma v=6\times {10}^{-27}{\mathrm{cm}}^3/\mathrm{s}$. In each case, the annihilation rate is normalized to a halo profile with $\gamma =1.3$. The point source spectrum is taken as the broken power law shown in Fig. 4, and the Galactic Ridge emission has been extrapolated from the higher energy spectrum reported by HESS [12], assuming a pion decay origin and a power-law proton spectrum. See text for details.

Figure 8

A histogram showing the distribution of spectral indices, $\Gamma$, of pulsars in the Fermi Pulsar Catalog.

Figure 9

The spectral indices (with statistical and systematic error bars) of the eight globular clusters observed by the Fermi Gamma Ray Space Telescope [54].

Figure 10

Upper limits on the dark matter annihilation cross section for several choices of the final state and for three values of the halo profile’s inner slope, $\gamma$. Also shown for comparison is the annihilation cross section predicted for a simple thermal relic ($\sigma v=3\times {10}^{-26}{\mathrm{cm}}^3/\mathrm{s}$). Uncertainties in the overall dark matter density have not been included, but based on the errors presented in Ref. 32, we expect that this would only weaken our limits by about 30–50%. See text for more details.

Figure 11

Upper limits on the dark matter annihilation cross section for several choice of the final state, neglecting the effects of baryons (using an uncontracted NFW halo profile). Also shown for comparison is the annihilation cross section predicted for a simple thermal relic ($\sigma v=3\times {10}^{-26}{\mathrm{cm}}^3/\mathrm{s}$). Uncertainties in the overall dark matter density have not been included, but based on the errors presented in Ref. 32, we expect that this would only weaken our limits by about 30–50%. See text for more details.